Biliary atresia (BA) is the most common cause of obstructive jaundice in infants and is lethal if uncorrected. BA affects 1/10,000 infants and is the leading indication for liver transplantation in children. The etiology of this fibroinflammatory disorder remains elusive, but recent work has implicated the cytokine interferon-gamma (IFN3) as having a crucial role, although the mechanism of IFN3 activation is unclear. IFN3 levels are elevated in patients with BA, and targeted deletion of IFNg inhibits development of the rhesus rotavirus (RRV)-injected mouse model of BA. The exclusive occurrence of BA in the developing liver suggests that developmental pathways play an important role in the pathogenesis of BA. Epigenetic modifications, such as DNA methylation, function as stable gene silencing mechanisms that repress gene expression in differentiated cells. Inhibition of DNA methylation, resulting in derepression of gene expression, has been shown to have a profound effect on development, and also increases the expression of IFN3 pathway genes. We have evidence that broad chemical (azacytidine [azaC]) or genetic (dnmt1, ahcy) inhibition of DNA methylation leads to defects in hepatobiliary development associated with a significant increase in IFN3 pathway genes, and that patients with BA demonstrate bile duct cell-specific DNA hypomethylation. These findings led us to hypothesize that bile duct cell DNA hypomethylation plays a pathogenic role in BA. We believe this to be mediated through TGF2, which plays a key role in biliary development and is inhibited by IFN3 activation. The goals of this proposal are to further understand the mechanisms by which inhibition of DNA methylation leads to developmental biliary defects, and to determine the importance of DNA methylation in the pathogenesis of BA. Our mechanistic studies will utilize zebrafish, in which we have considerable experience examining hepatobiliary development. Zebrafish are a facile model organism for studying development, as they develop rapidly ex vivo and there are considerable tools available for genetic manipulation and analysis. We will also utilize mouse models and take advantage of the high volume of patients with infantile biliary disease at The Children's Hospital of Philadelphia. Using these resources, we propose to test the hypothesis that DNA hypomethylation leads to biliary defects in zebrafish and mouse models and in patients with BA, and that activation of IFN3 plays a critical role.